Oral and Enteral Resuscitation of Burn Shock The Historical Record and Implications for Mass Casualty Care

The Importance of Early Resuscitation and the Need for Alternatives to IV Therapy

Fluid resuscitation must be started promptly in order to achieve the best outcomes. The high survival rates of modern burn care are attributed to application of prompt and aggressive resuscitation using intravenous infusion of lactated Ringer's.¹² Half or more of the fluid requirements in the first 24 hrs must be administered in the first 8 hrs for optimal outcomes.^13,14 Studies have shown that severely burn-injured children have increased mortality rates and higher rates of organ failure when initiation of fluid therapy is delayed by even a few hours.^12,15

There is a need to have a better means for the initial resuscitation of burn shock. Most of the world's population does not have access to prompt advanced medical care and IV therapy in the event of severe injury. Burn resuscitation can be delayed in rural regions, in underdeveloped countries, and in a variety of circumstances that delay notification and dispatch or limit medical resources. Initiation of IV access can be particularly delayed in children because of difficulties with cannulating small vessels through burned tissue. At the Galveston Shriners Hospital, referrals of burn-injured children often have a record of delayed resuscitation.

Mass disaster in which the large number of casualties simply overwhelm care providers is another situation where intravenous fluid therapy for the treatment of burn injury will necessarily be delayed. Perhaps at no time in our country's history have we been more aware of the potential for mass casualties; we must plan for the best means to treat thousands of victims of explosions and fires. The September 11, 2001 terrorist attacks caused explosions and fires that resulted in scores of trauma and burn-injury casualties that were effectively treated at New York and Washington trauma and burn centers. However, the medical challenge could have been far greater. If the attacks had come a few hours later or if the World Trade Center towers had not collapsed, there could have been thousands of casualties with large and severe burns requiring prompt fluid resuscitation. Doctors at New York and Washington burn centers contemplated the impossibility of providing prompt resuscitation when dealing with such a number of patients.¹⁶

Fluid therapy is also a critical component of combat casualty care. Acute hemorrhage is the major cause of death in conventional warfare, accounting for about 50% of all battlefield fatalities as reviewed and analyzed by Bellamy.¹⁷ Bellamy also noted that administration of effective first aid and resuscitation solutions in the field could reduce battlefield mortality by ~20%. Unfortunately, fluid therapy for combat casualty care can be delayed, limited, and inadequate owing to logistical constraints for field personnel and supplies and delays in casualty evacuations due to tactical situations.

Preclinical Studies of Enteral Resuscitation of Burns and Trauma

Sanford Rosenthal was a research pharmacologist at the NIH who conducted an extensive series of experiments dealing with experimental treatment of burns and trauma.^2,3,19 Most of the research was done using mouse models of hemorrhage, burn injury, or traumatic shock (tourniquet). Rosenthal's shock model was designed to have a high mortality rate in untreated controls (typically >50% in less than 6-hrs). A consistent finding of these studies was the great survival benefit of large volumes of isotonic NaCl when administered via the oral or intraperitoneal routes. Pure water offered virtually no benefit and 5.5% glucose solutions offered little benefit compared to untreated controls. Effectiveness of plasma or saline administered intravenously was demonstrated, but generally IV infusions were not as effective as that of large volumes of oral saline equal to 100-150 mL/kg for treatment of a 100% TBSA partial-thickness burn and 80 mL/kg for treatment of a 40-50 mL/kg hemorrhage.^2,3

It is surprising that Rosenthal reported that equal volumes of IV saline or plasma were often less effective than enteral administration of saline, but this may reflect that all fluids appear to have been administered as a bolus. An oral (gastric) bolus would enter the circulation over time, while an IV bolus might cause acute hemodynamic overload. We can imagine the cardiovascular effects of a 50-100 ml/kg intravenous bolus that is equal to 1–2 blood volumes. Thus, it is impossible to extrapolate Rosenthal protocols and results against modern standard of care treatments. Nevertheless, the importance of Rosenthal's studies is that they demonstrated an efficacy of gastric infusion and enteral resuscitation in severe shock due to burns, hemorrhage, or trauma, and that they led to several clinical trials.

Plasma was established as the vascular deficit of burn shock in the mid-20th century and this led to the widely accepted belief at the time that intravenous infusion of plasma was the best means to reverse the sequelae of burn shock^31-33 However, evaluations of treating mass casualties at Pearl Harbor and the Coconut Groove fire left burn doctors unsatisfied with the mortality of major burns.⁶ There were few basic science studies at this time, but the studies published on burn resuscitation focused on the value of plasma vs saline. Rosenthal,^3,19 and Baxter and Shires^34,35 were pioneers when they suggested that large volume resuscitation with crystalloids, specifically saline or LR, was more beneficial than plasma for treatment of burns and hemorrhage.

Clinical Studies Using Oral and Enteral Resuscitation of Severe Burn Injury

Enteral resuscitation of moderate to severe burn injury has been described in twelve studies that we have identified; listed here and in Table 1, Fox (1944), Moyer (1949), Markley (1956), Wilson and Stillman (1960), Davies (1964), Franke (1964), S0rensen (1968), Jackson (1966), Monafo (1970), Ahnefeld (1975), Maksimov (1989) and El-Sonbaty (1991).^{5,6,10,11,20-26,36} Despite this background few surgeons or intensivists consider oral or enteral resuscitation as even partially supportive for treatment of severe hypovolemia. Results are summarized in Table 1 and discussed to include both benefits and complications of each clinical study of enteral resuscitation of burn injury that we have found. The patient numbers (n=) refer only to patients treated with enteral resuscitation and do not include numbers from control groups. We found no published clinical studies on enteral resuscitation of trauma or hemorrhagic hypovolemia.

In 1926 Davidson reported how burn patients in his day often had hyponatremia and hypochloremia.³⁷ Subsequently, he reported efficacy with large, as much as 10 liters, infusions of isotonic saline administered in first 24-hrs by three routes, subcutaneous, intravenous and rectal.

The first published study we identify describing the exclusive use of oral or enteral resuscitation of severe burns was by Charles Fox in 1941. Fox used oral resuscitation treatment exclusively for both partial and full thickness injuries, but he reported details on nine severely injured patients with moderate to large full thickness injuries. Four children (TBSA 23%–80%) and five adults (19%–41%) with full thickness burns and who exhibited signs of shock were treated by Fox with chilled isotonic sodium lactate given upon admission and thereafter at 15 minute intervals to administer a total 100-150 mL/kg in 24-hrs.^6-8 As calculated per the Parkland formula, this volume would be considered sufficient by current standards to resuscitate a 25–38% TBSA injury. Vomiting, which occurred frequently in severe burns, was treated by administration of more fluid and in these patients nasogastric tubes were used to provide continuous fluid delivery. Fluid infusions were adjusted to obtain 1–2 liters of urine per day. When food was tolerated, a high protein diet was started.

Fox reported “In general the large volumes of sodium lactate were well tolerated, the patients wanted water to drink, but after a short time became accustomed to the lactate and drank copiously of their own volition.”

In 1949, Moyer reported on his approach to oral resuscitation used to treat more than 30 severely burned children and adults.⁵ Moyer found that NaCl alone caused an acidosis, which could be prevented with bicarbonate, lactate, or citrate. These buffered and balanced saline solutions were also more palatable and induced fewer indices of nausea and vomiting. Moyer suggested that the citrated formula was perhaps the most palatable. Moyer was careful to state that when circulatory collapse was present, it was necessary to administer IV fluid with lactated Ringer's or plasma until the peripheral collapse was corrected after which oral resuscitation was initiated. Moyer did not present any quantitative guidelines to determine at what level of shock enteral resuscitation is contraindicated and at what level it becomes effective.

In 1956, Markley et al. reported on the results from an extensive NIH sponsored trial conducted in Peru that evaluated the use of oral resuscitation. This study was one of the only controlled trials of oral resuscitation in which half the patients received IV plasma and blood per the standard of care of his day. Treatment was alternated for successive patients between oral saline or IV therapy using plasma and saline. Markley used an isotonic bicarbonated saline administered solely or chiefly orally as the emergency treatment in the first 48-hr for severe burns. Entry criteria were TBSA > 10% with mean burn size being ~40% and ~30% in 55 children and 56 adults, respectively. Control groups of comparable size were administered IV plasma or blood, and some saline with volume determined per the Evans formula. By modern standards we would consider the 1-ml/kg volume of the Evans formula to be inadequate volume.

The volume of oral saline was 110 mL/kg in the first 24-hrs. The 48-hr mortality data were equal or better with oral resuscitation, while data on hematocrit and urine output suggested effective volume expansion and improved hemodynamics with oral resuscitation.

Markley's mortality data with oral resuscitation was similar to standard of care mortality results reported from U.S. urban burn centers of the day. However, urine output and mortality data grouped for different size burns suggest an inability to fully resuscitate patients with burns >50% TBSA. The dose of fluid administered was independent of burn size, 110 mL/kg during the first 24-hr. This is a volume equivalent to the Parkland formula for a 27% TBSA injury. Thus, we could expect that inadequate volumes may have been administered to Markley patients with burns larger than 30% TBSA. Two unanswered questions are: would outcomes be improved with oral administration of larger volumes, and what is the limit on volume that can be delivered to a severely burn injured patient via the enteral route?

In 1960, Wilson and Stirman reported on outcomes of 142 burn-injured patients with 15-60% TBSA injuries treated with isotonic saline compared to saline plus blood.²¹ Of course, blood was administered via IV route. Mortality rate was higher in the group that received blood and Wilson made the case that blood was rarely needed for burn resuscitation. All patients were treated with oral intake of the salt solution with IV being used initially only if the patients demonstrated peripheral circulatory collapse, defined as low or no blood pressure. When oral treatment seemed inadequate to maintain the circulation or impractical because of vomiting, these patients were administered IV lactated Ringer's solution or plasma until stabilized. However, 81 of Wilson's patients were treated exclusively with oral intake including some patients with 60% TBSA injury. It is unclear if Wilson's patients may have also received some nutrition that would have provided glucose or other absorbable solutes. Most subsequent clinical trials of enteral resuscitation reported on the use of oral delivery of bicarbonated, citrated, or lactated saline solutions that were more palatable, and associated with less nausea and vomiting.

Davis reported in 1964 for the British Medical Research Council on the increased use of IV plasma for burns that was made available by Britain's National Blood Transfusion Service (NETS). The NETS was launched during the London Blitz in World War II, and to this day provides British hospitals with an adequate supply of low-cost blood products. The focus of the Davis study was on isotope measurements of RBC volume and plasma volume in burn patients, 10-95% TBSA, subjected to routine care which included oral water and oral Moyer solution administered at 2x normal oral fluid input, a rate independent of injury size. Patients with burns of 40% TBSA or less were resuscitated initially with oral crystalloids. This volume was clearly inadequate in several patients as half of these required subsequent IV saline or saline plus plasma, started when vomiting was excessive or clinical condition deteriorated. On the other hand, in half of the patients oral resuscitation was adequate with good clinical outcomes.¹⁰

Franke and Kock-Marburn reported in 1964 on the first clinical use of oral electrolyte solution that also contained glucose for treatment of severe burn injury in children in the German language journal, Langenbeck Arch Klin Chir.²² This group often exclusively used gastric infusion and an anti-emetic to treat 19 out of 22 children with 8 to 38% TBSA injuries, all of whom survived. One 2 year old with a 31% TBSA burn including 7% full thickness received 3,000 mL and 2,300 mL enterally on day 1 and day 2, respectively. Three additional children with larger burns of 35 to 70% received IV fluid and gastric infusions and two of these died from cerebral edema and heart failure.

In 1966, Jackson and Chen¹¹ set out to determine if a moderate sized burn could be safely treated enterally with Meyer's solution, since most burn centers advocated IV plasma. Out of 162 burn cases (113 children and 49 adults) with 10% to 35% TBSA injuries, 75% were exclusively treated with oral resuscitation, 17% received IV LR infusions and 8% “required” IV colloid infusions. Vomiting occurred in 30% of the children and 51% of the adults and thus vomiting was not necessarily the criterion to stop oral resuscitation or start IV therapy. Mean urine output was reported less than normal in 73% of the patients at 8-hrs postburn, with this number dropping to 44% at 24-hours and 34% at 448-hours. Children had a substantially lower percentage of oliguria than in adults. This study suggests better toleration and efficacy of enteral resuscitation in children compared with adults. Similar conclusions were recently reached by Brown et al. who showed younger children have less vomiting with enteral feedings.³⁸

Based on the published levels of urine output in most pre-1980 studies, it is apparent that severely injured (>50% TBSA) burn patients were often under-resuscitated compared to our current standards. Burn patients almost always requested water to drink and while not a stated part of clinical care, drinking water was allowed in early burn care. This was often a mistake as significant water consumption without sodium supplementation was shown to cause an early “toxemic phase of burn injury” due to water intoxication.⁵ Paracelsus (1493-1541) noted that severe burn injury was accompanied by fever and a ‘drying out of the blood’, resulting in thirst. However, he recommended that if a patient drinks, death will follow. Drinking Na free fluids can cause severe hyponatremia and cerebral edema. On the other hand, drinking electrolyte solution is an effective means of burn resuscitation. Animal and clinical studies both showed the critical value of replacing oral water with oral salt solutions.^2,5,6,20

A novel approach to oral resuscitation that might allow safe drinking of non-sodium-containing beverages was suggested by Sørensen who sought a means to treat mass casualties due to thermal nuclear warfare. The impetus for his work was the U.S.-Cuban missile crisis of 1962. Subsequently, Sørensen resuscitated 26 consecutive patients by allowing them to drink any beverage they wanted, which reportedly was tea, beer, milk, or water. Sørensen reported that salt solutions had a disagreeable taste and thus administered 7.5-g salt tablets 5-g NaCl and 2.5-g NaHCO3 per each liter of fluid. Similar salt tablets have been commercially available, but currently are difficult to find. Sørensen setup an IV drip if needed, but most of his patients did not require it. Children and adults could take the tablets. Fluid intake was not controlled per burn size and totaled 150 to 200 mL/kg in first 24-hrs and 100 to 150 mL/kg in the second 24-hrs. Eighty percent of these patients required no IV support and this group had burns as large as 45% TBSA. To our knowledge, this unique approach to mass casualty care was never followed up or used again.

In 1970, Monafo reported on resuscitation after major burn injuries using 600 mOsm hypertonic lactated saline (HLS) in which patients were orally and intravenously resuscitated.²⁴ Monafo used HLS to partially resuscitate 4 burn adults with 30% to 95% TBSA burns and 3 children with 22% to 58% TBSA burns. Most were full thickness injuries. Patients received 10-60% of their fluid needs in the first 24 hrs from oral fluids and 60-90% of their fluid needs in the second 24 hrs. This report shows that partial oral resuscitation of severe burns can be successful and that it can be accomplished with a hypertonic solution without glucose. In subsequent years Monafo continued to use IV hypertonic saline, but his use of oral resuscitation was not continued. To our knowledge, Monafo's suggestion for efficacy of oral hypertonic saline has never been followed-up either clinically or experimentally.

Rosenthal had found hypertonic saline to be less effective than isotonic saline treatment of burn-injured mice. Other early studies reported that hypertonic saline was less palatable and led to greater incidences of vomiting. However, Monafo stated, “it was discovered that HLS solution (iced to be made palatable) was easily tolerated orally in significant amounts without apparent GI dysfunction.” One wonders if the ice not only chilled, but also significantly diluted the concentration of the isotonic saline. Hypertonic fluid that enters the intestine rapidly becomes isotonic due to mobilized fluid, but thereafter can be rapidly absorbed along with its sodium, particularly if mucosal blood flow is enhanced as has been reported after hypertonic resuscitation using IV infusion.³⁹

In 1975, Ahnefeld reported on the use of Liquidsorb, a slightly hypertonic 370 mOsm electrolyte ‘lemonade’, for treatment of 68 burn patients with 12 to 34% TBSA injuries.³⁶ Liquidsorb contained 4 g of glucose, 100 mg ascorbic acid, and 12 mg nicotinamide per liter of bicarbonated saline. Only 4 of 58 who arrived at the hospital within one hour of injury could not drink the solution. However, 16 of 23 patients who arrived more than 2-hours after injury were not able to drink because of “sickness, vomiting, centralization of the circulation, and unconsciousness.” These data suggest that in order to be effective, oral resuscitation must be started early, ideally within the first hour after injury.

Ten subsequent burn patients with 12 to 26% injuries were also treated with Liquidsorb and demonstrated good absorption of one liter in the first hour and 750-mL in the second hour when delivered by stomach tube in 50-mL increments. This approach was well tolerated except for patients that had “centralization” or peripheral vascular failure as evident by collapsed peripheral veins and delayed capillary refill. However, oral resuscitation was effectively implemented in patients that were initially resuscitated with IV therapy and after reestablishment of “normal circulatory conditions”.

Modern Studies of Enteral Resuscitation of Burn Shock

Surgical textbooks and the Advance Burn Life Support (ABLS) course suggest that oral fluids are an adequate means to treat volume losses in small burns of 5-10% TBSA. Recently, Brown et al have suggested considering enteral resuscitation on larger burns up to 40% TBSA alone or used in conjunction with IV fluids.³⁸ We suggest that TBSA should not be the sole criterion for the appropriateness of enteral resuscitation. Larger but less severe burns should do well with enteral resuscitation. While smaller burns with deep injury, particularly with additional trauma, could require IV fluid therapy. There is no consensus on how best to perform enteral fluid therapy with small burns. A recent survey of burn centers in the United Kingdom shows a wide divergence of fluids used for oral resuscitation of 5% to 10% TBSA burns including electrolyte solutions and beverages (i.e. water, milk and juice).⁴⁰

We could identify only a few publications in the last 20 years on enteral resuscitation of large burns and none were in Medline referenced literature. El-Sonbaty reported good results with oral resuscitation of 20 children with 10-20% TBSA burns using the WHO ORS.²⁶ The depth of the burns was not presented, but we can assume that most of the patients did not present in burn shock as the children's mothers under the supervision of the nurse administered the oral hydration. The volume and rate of oral hydration with WHO ORS used by El-Sonbaty was identical to standard Parkland formula for IV administration. He reported oral hydration to be as effective as IV lactated Ringer's, also administered at Parkland rate. Again there is no way to accurately compare relative vascular expansion or effectiveness on the two groups. Burn injuries of this small size do not induce hypovolemic shock. Patients in both groups developed hyponatremia (125-130 mEq/1) on day 5 post-injury, but otherwise had unremarkable outcomes. A comparison of compositions of oral hydration solutions versus IV solutions shown in Table 2 suggests that hyponatremia may be a problem with most oral solutions and that oral resuscitation of burned patients may require a solution with a higher sodium concentration, such as used by Monafo.

Jiang reported the only animal study of oral resuscitation of severe burn injury that we have found in which the control group was treated with the modern Parkland formula.¹⁸ Anesthetized dogs were inflicted with a 30% partial thickness burn and then treated with a 347 mOsm “burn drink” of glucose NaCl and NaHCO3 (Table 2). Total volume administered over 24 hrs was the Parkland formula, 4 ml/Kg per % TBSA. Controls were untreated burns, and burns treated with a 1/10 dilution of the burn drink (35 mOsm). Impressive improvements in plasma volume, cardiac output and urine output were shown, but only for the 347 mOsm “burn drink” group, not for the hypotonic enteral group.

Despite a fairly extensive literature on enteral resuscitation of burn shock, most present-day clinicians are unaware of its utility as an option, except in minor burns where resuscitation is not critical. However, two modern clinical trends speak to the question of efficacy and safety of enteral resuscitation. These are oral hydration in the treatment of cholera, and aggressive enteral feeding of trauma patients.

Treatment of cholera

Oral hydration has been a focus of basic and clinical research for the treatment of life threatening dehydration due to diarrhea. Dehydration and associated shock due to cholera and diarrhea are the leading causes of pediatric mortality in the world.⁴¹ There have been dramatic reductions in the mortality rate of cholera victims in underdeveloped countries due to oral rehydration solutions that are easily administered by the patient's family or other lay persons. The successful use of glucose-salt solutions such as the WHO ORS has been called one of the major medical advances of the 20th century based on the tens of millions of lives saved.⁴¹ Hydration with pure water is an ineffective treatment of cholera, but providing substrate for the intestine's coupled sodium-glucose transporter provides a means to replace electrolytes, energy substrate, and water in severe dehydration. Specifically, WHO ORS packets provided by the World Health Organization and governments have provided a low cost and highly effective treatment of dehydration due to cholera. Children and their mothers in under-developed countries are very familiar with the WHO ORS packets and how to make up the solutions and administer them to treat dehydration due to diarrhea. Most trained specialists in the U.S. are unfamiliar with WHO ORS and treat dehydration exclusively with IV infusions.

Early enteral feeding of burn injury

Recent research has established significant clinical advantages of early intestinal or enteral resuscitation even if it can only be performed in hospitals. Enteral feeding may protect the gut by enhancing or maintaining the mucosal barrier, which can be a route for systemic entry of bacteria and endotoxins. While early enteral feeding has been traditionally avoided in burn patients, it has recently been demonstrated to be safe and effective for nutrition when started immediately with hospital care of patients with large burns⁴² or used before, during, and immediately after surgery of burned patients.¹⁸ The gut benefits from early enteral resuscitation, particularly when specific nutrients are included. Enteral resuscitation increases intestinal blood flow and results in better maintenance of gut barrier integrity. This may reduce the risk of sepsis and multi-organ failure.⁴³

Composition of Oral Hydration Solutions

The optimal composition of oral replacement fluid has not been determined and will likely vary for different indications. The World Health Organization has had great success with slightly hypertonic ORS solution containing glucose, sodium, chloride, and bicarbonate. It is easily formulated by adding WHO ORS powder packets to water, Table 2. Gatorade®, PowerAde® and other “sports drinks” are low in sodium (5-20 mmol/L), as they are designed to replace perspiration losses, as well as to increase palatability.⁴⁴ Monafo (1970) utilized a 600 mOsm hypertonic lactated saline solution (HLSS) containing sodium, racemic D-L lactate and chloride to orally resuscitate 10 burn injury patients.²⁴ It seems logical that a solution similar to Ringer's lactate or HLS with the addition of glucose with an osmolarity range of 260-330 mOsm/L given by drinking or infusion through a nasogastric tube could rapidly be absorbed and provide the large volume and Na necessary for burn resuscitation. Although such a formulation appears logical, it will require substantial research to define an optimal solution.

Sports Drinks

Oral hydration in developed countries has not focused on treatment of disease, but rather on sports drinks, such as Gatorade, to provide effective hydration of people engaged in strenuous activities. The financial success of the sports drink industry with sales of 1.5 billion dollars per year has stimulated both research and the creation of numerous marketed drinks for hydration and nutrition. The hydration effect of such drinks is based on the same principles as those of the WHO ORS.^45,46 The higher sodium concentration of WHO ORS versus Gatorade reflects the different electrolyte losses associated with diarrhea and sweating, respectively. Sweat has a sodium concentration of 30-60 mEq/L, which is less than the 60-120 mEq/L in watery stools or of the 130 mEq/L of plasma extravasated into a burn wound. Resuscitation of severe hypovolemia as with burn shock is a greater challenge than dehydration due to sweating or diarrhea because effective solutions used for volume expansion, such as LR, require a higher sodium concentration than most modern oral hydration formulas and could require delivery of larger volumes. On the other hand, Table 1 shows that both isotonic and hypertonic solutions without glucose and with higher sodium concentrations were effective for resuscitation of severe burn shock.

Oral Hydration to Expand Astronaut's Blood Volume

NASA has employed different oral hydration strategies for astronauts prior to shuttle reentry to expand blood volume and prevent orthostatic hypotension during descent. Oral rehydration with salt water or with water and salt tablets was shown to expand blood volume as a counter measure against the loss of vascular volume and cardiovascular tone that occurs during prolonged space flight. NASA's Greenleaf et al studied formulations with differences in osmolarity and sodium concentrations.⁴⁷ They concluded that cation content is more important than osmotic content for plasma volume expansion. This led to the development of AstroAde a high-sodium oral hydration solution (Table 2). AstroAde was effective at expanding plasma volume of dehydrated subjects during exercise.⁴⁸ Table 2 compares the compositions of selected oral hydration solutions and IV solutions.