Jeffrey R. Saffle

American Burn Association consensus conference to define sepsis and infection in burns.

Because of their extensive wounds, burn patients are chronically exposed to inflammatory mediators. Thus, burn patients, by definition, already have “systemic inflammatory response syndrome.” Current definitions for sepsis and infection have many criteria (fever, tachycardia, tachypnea, leukocytosis) that are routinely found in patients with extensive burns, making these current definitions less applicable to the burn population. Experts in burn care and research, all members of the American Burn Association, were asked to review the literature and prepare a potential definition on one topic related to sepsis or infection in burn patients. On January 20, 2007, the participants met in Tucson, Arizona to develop consensus for these definitions. After review of the definitions, a summary of the proceedings was prepared. The goal of the consensus conference was to develop and publish standardized definitions for sepsis and infection-related diagnoses in the burn population. Standardized definitions will improve the capability of performing more meaningful multicenter trials among burn centers.

Recent outcomes in the treatment of burn injury in the United States: a report from the American Burn Association Patient Registry.

This article reports outcomes of 6417 patients treated during 1991-1993 in 28 burn centers. Data were gathered by use of the ABAs computerized patient registry. Mean burn size was 14.1% total body surface area. There were 6087 survivors (95.9%); mortality rate among patients with inhalation injury was 29.4%. A probit analysis was used to construct a series of survival curves; the LA50 (which is defined as the burn size lethal to 50% of patients) for young adults was 81% total body surface area. Mean length of hospital stay was 13.5 days. Total mean charges were

Multicenter postapproval clinical trial of Integra dermal regeneration template for burn treatment.

39,533, with resource utilization related to clinical comorbidity factors and length of stay. No reliable method could be found to predict survival on admission. Mean charges for the most intensive diagnosis-related group (No. 472) exceeded

The Phenomenon of "Fluid Creep" in Acute Burn Resuscitation

198,000 per patient. Pooled data and related summary statistics used in this study have some limitations, which are discussed. Additional accurate information regarding the outcomes of thermal injury treatment is needed by the burn care community.

Effect of inhalation injury on fluid resuscitation requirements after thermal injury

The safety and effectiveness of Integra Dermal Regeneration Template was evaluated in a postapproval study involving 216 burn injury patients who were treated at 13 burn care facilities in the United States. The mean total body surface area burned was 36.5% (range, 1-95%). Integra was applied to fresh, clean, surgically excised burn wounds. Within 2 to 3 weeks, the dermal layer regenerated, and a thin epidermal autograft was placed. The incidence of invasive infection at Integra-treated sites was 3.1% (95% confidence interval, 2.0-4.5%) and that of superficial infection 13.2% (95% confidence interval, 11.0-15.7%). Mean take rate of Integra was 76.2%; the median take rate was 95%. The mean take rate of epidermal autograft was 87.7%; the median take rate was 98%. This postapproval study further supports the conclusion that Integra is a safe and effective treatment modality in the hands of properly trained clinicians under conditions of routine clinical use at burn centers.

Multiple organ failure in patients with thermal injury.

Several reports have documented that modern burn patients receive far more resuscitation fluid than predicted by the Parkland formula—a phenomenon termed “fluid creep.” This article reviews the incidence, consequences, and possible etiologies of fluid creep in modern practice and uses this information to propose some therapeutic strategies to reduce or eliminate excessive fluid resuscitation in burn care. A literature review was performed of historical references that form the foundation of modern fluid resuscitation, as well as reports of fluid creep and its consequences. The original Parkland formula required a 24-hour volume of 4 ml/kg/%TBSA lactated Ringers solution followed by an infusion of 0.3–0.5 ml/kg/%TBSA plasma. Modern iterations of this formula have omitted the colloid bolus. Numerous exceptions to the formula have been noted, most consistently patients with inhalation injuries. In contrast, recent reports document greatly increased fluid requirements in unselected patients, which seems to consist largely of progressive edema formation in unburned areas, increasing after the first 8 hours post-burn. This has been linked to occurrence of the abdominal compartment syndrome and other serious complications. Strategies to reduce fluid creep include the avoidance of early overresuscitation, use of colloid as a routine component of resuscitation or for “rescue,” and adherence to protocols for fluid resuscitation. Fluid creep is a significant problem in modern burn care. Review of original investigations of burn shock, coupled with modern reports of fluid creep, suggests several mechanisms by which this problem can be controlled. Prospective trials of these therapies are needed to confirm their effectiveness.

Plasmapheresis as an adjunct treatment in toxic epidermal necrolysis.

The presence of inhalation injury has been reported to increase fluid requirements for resuscitation from burn shock after thermal injury. To evaluate the effect of inhalation injury on the magnitude of burn-induced shock, the characteristics of resuscitation of 171 patients with burns covering at least 25 percent of the total body surface area were reviewed. When inhalation injury was suspected, confirmation by xenon-133 scanning, bronchoscopy, or both was obtained. Initial fluid resuscitation was calculated according to the Parkland formula, and titration was initiated to maintain a urine output of 30 to 50 ml/hour. Fifty-one patients had inhalation injuries. Patients with inhalation injuries had a mean fluid requirement of 5.76 ml/kg per percentage of total body surface area burned and a mean sodium requirement of 0.94 mEq/kg per percentage of total body surface area burned to achieve successful resuscitation, compared with a fluid requirement of 3.98 ml/kg per percentage of total body surface area burned and a sodium requirement of 0.68 mEq/kg per percentage of total body surface area burned for the group without inhalation injury (p < 0.05). These data confirm and quantitate that inhalation injury accompanying thermal trauma increases the magnitude of total body injury and requires increased volumes of fluid and sodium to achieve resuscitation from early burn shock.

Effects of oxandrolone on outcome measures in the severely burned: a multicenter prospective randomized double-blind trial.

Objective:To assess the frequency and significance of multiple organ failure in patients with burn injuries. Design:Retrospective review and prospective assessment of patients with acute burns. Setting:University hospital burn center. Patients and Methods:We reviewed 529 patients admitted for acute burn treatment whose lengths of stay exceeded 72 hrs. A new scoring system, the Thermal Injury Organ Failure Score, was used to assign scores from 0 (normal) to 6 (severe dysfunction) to each of 6 organ systems, which were then totaled to compile the overall score. This system was also used for prospective assessment of 83 adult burn patients, and compared with the Acute Physiology and Chronic Health Evaluation (APACHE) II scoring system during the first week of treatment. Interventions:None. Measurements and Main Results:For 496 survivors, mean organ failure score was 3.28, compared with 23.1 in 33 nonsurvivors (p < .0001). All nonsurvivors but one had scores of ≥15, indicating dysfunction of at least three organs. Scores and mortality rate increased with age and burn size. Pulmonary dysfunction was the most frequent form of organ failure seen, but correlated less with outcome than did cardiovascular or neurologic scores. Sepsis was present in 22 of 33 patients who died. In the prospective study, organ failure scores correlated with outcome more closely than did APACHE H scores. Weekly evaluation of these patients demonstrated progressive divergence in scores between survivors and nonsurvivors. Conclusions:Multiple organ failure was almost invariably present in burn patients who died >72 hrs after injury. Burn victims, who have been excluded from reviews of multiple organ failure, appear to manifest organ failure in a manner similar to that of other surgical populations. The scoring system reported here may prove useful in evaluating organ failure in thermally injured patients. (Crit Care Med 1993; 21:1673–1683)

National Burn Repository 2005: a ten-year review.

BACKGROUND Toxic epidermal necrolysis (TEN) is a severe, progressive disease characterized by the sudden onset of skin necrosis. It is frequently associated with systemic involvement and has a high rate of morbidity and mortality. Standard therapy includes meticulous wound care, fluid replacement, and nutritional support in an intensive care setting. OBJECTIVE We evaluated the outcomes of patients treated in a burn unit for TEN over a 9-year period and compared the outcomes of a subset of patients treated with plasmapheresis with those managed by conventional means. METHODS The records of 16 patients with a diagnosis of TEN obtained from a computerized database were reviewed. Parameters recorded included extent of body surface area involvement and number of mucous membranes involved at admission, complications such as sepsis or need for mechanical ventilation, length of stay, and disposition. RESULTS Sixteen patients were included in this study. Ten were treated with conventional support measures alone. Six were treated with plasmapheresis. The average age was 42.4 years; the male/female ratio was 1:2.2. Sulfamethoxazole/trimethoprim was implicated in causation in 6 patients. The average extent of involvement on admission in all patients was 51.5% total body surface area. The average length of stay in all patients was 14.8 days. Eight patients (50%) were discharged home, 4 (25%) were discharged to a rehabilitation facility, and 4 (25%) died (2 of sepsis, 2 of cardiopulmonary arrest). None of the plasmapheresis-treated patients died. CONCLUSION Plasmapheresis is a safe intervention in extremely ill TEN patients and may reduce the mortality in this severe disease. Prospective studies are needed to further define its usefulness.

Etiology and consequences of respiratory failure in thermally injured patients

Severe burns induce pathophysiologic problems, among them catabolism of lean mass, leading to protracted hospitalization and prolonged recovery. Oxandrolone is an anabolic agent shown to decrease lean mass catabolism and improve wound healing in the severely burned patients. We enrolled 81 adult subjects with burns 20% to 60% TBSA in a multicenter trial testing the effects of oxandrolone on length of hospital stay. Subjects were randomized between oxandrolone 10 mg every 12 hours or placebo. The study was stopped halfway through projected enrollment because of a significant difference between groups found on planned interim analysis. We found that length of stay was shorter in the oxandrolone group (31.6 ± 3.1 days) than placebo (43.3 ± 5.3 days; P < .05). This difference strengthened when deaths were excluded and hospital stay was indexed to burn size (1.24 ± 0.15 days/% TBSA burned vs 0.87 ± 0.05 days/% TBSA burned, P < .05). We conclude that treatment using oxandrolone should be considered for use in the severely burned while hepatic transaminases are monitored.