Predicting Death in Burned Patients
ABSTRACT & COMMENTARY
Synopsis: The probability of death following burn injury can be predicted through the use of three risk factors readily identifiable on admission.
Source: Ryan CM, et al. N Engl J Med 1998;338(6): 362-366.
Records of all 1665 patients admitted to two large burn centers from 1990 to 1994 were reviewed to identify variables that might aid in distinguishing between those patients who will survive and those who will die. The investigators tallied the patients' age, gender, length-of-stay, type and extent of burn, presence of inhalation injury, need for escharotomy, and mortality. Inhalation injury was considered to be present if the fire occurred in a closed space, if bronchoscopy revealed lower airway soot, or if the blood carboxyhemoglobin level was elevated on admission.Mean age of the 1665 patients was 21 ± 20 years, with a range of 1 month to 99 years; 69% were male. Mean burn size was 14 ± 20 percent of body surface area (total of second- and third-degree burns). Fifteen percent of the patients had inhalation injury. Mean length of hospital stay was 21 ± 29 days. Sixty-seven patients died (4%).
Three risk factors were predictive of mortality: age older than 60 years, body surface area of burn larger than 40%, and the presence of inhalation injury. The investigators developed a mortality formula using these three factors to predict mortalities of 0.3%, 3%, 33%, and roughly 90%. This mortality formula was then applied using admission data on a second, prospective series of 530 patients admitted in 1995 and 1996. The three risk factors proved to be highly predictive of mortality in the prospective patient series, as summarized in the table. Ryan and colleagues conclude that the probability of death after burns is low, considering the totality of patients admitted to a major burn center, and that mortality can be predicted soon after injury on the basis of simple clinical criteria.
Table
Actual and Predicted Mortality According to Number of Risk Factors
PredictedNumber of Number of Number of number Actual Predicted
risk factors patients deaths of deaths mortality mortality
0 1314 3 3 0.2% 0.3%
1 218 10 8 5.0% 3.0%
2 111 33 37 30.0% 33.0%
COMMENT BY DAVID J. PIERSON, MD
When I started practice, a reasonable percentage estimate of whether a given patient would succumb to a burn injury could be obtained using the patient's age in years plus the percentage of body-surface area burned. However, the tremendous improvements in survival following burn injuries that have occurred during the last 20 years render this and other traditional mortality predictors obsolete. Other recent authors have derived prediction formulas for mortality, although these have tended to be complex and difficult for individual clinicians to apply at the bedside.This study shows that, at least for the population of burned patients seen at Massachusetts General Hospital and the Shriners Burn Institute in Boston, mortality can be estimated with considerable accuracy using three features readily available on admission. These factors are age older than 60, total burned area larger than 40%, and the presence of indicators of inhalation injury. Mortality in the absence of any of these factors is 0.2-0.3%; in the presence of any one factor this is increased 10-20-fold; with two factors, the likelihood of mortality is about one-third, and with all three, approximately 90%.
There are potential problems with this study. It was done in two burn
centers within a single hospital system where a uniform approach to burn
care was used that might not correspond to other clinical settings. There
were relatively few deaths, and the criteria for the presence of inhalation
injury might not be accepted everywhere. Unusual burn injuries (e.g., a
relatively young patient with a near-total body burn but without inhalation
injury) as might occur with scalding, may not be relevant to the predictor,
as Ryan et al point out in the discussion. Still, the predictors developed
in this study have the advantages of being readily available and easy to
apply. Further experience in a variety of clinical settings will determine
the clinical applicability of these predictors and help to identify any
exceptions or needed modifications.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.